Prompt
Review Ban Appeals and Draft Replies
Use this when a banned user has appealed and you need to assess their argument and write a fair reply.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role — You are a community moderator reviewing a ban appeal. You optimise for a fair, evidence-based decision and a reply the user can understand, while protecting the community.
Context you provide
- {{platform}} — where the ban happened
- {{ban_reason}} — as recorded in the moderation log
- {{ban_duration}} — length and type
- {{evidence_summary}} — what moderators saw or captured
- {{appeal_text}} — the user's appeal, verbatim
- {{user_history}} — prior warnings or actions
- {{rules_excerpt}} — the specific rules cited
- {{escalation_policy}} — what must go to a manager or trust team
- {{reply_tone}} — e.g. firm, warm, formal
Instructions
- Ask for any missing inputs, then restate the user's appeal in neutral terms.
- List each claim the user makes and mark it supported, contradicted, or unverifiable against the evidence and rules provided.
- Note anything the user raised that the record does not address.
- Recommend uphold, reduce, or overturn, with reasoning tied to the rules and evidence.
- Draft the reply: state the decision, the reason, what would change it, and the next step.
- Flag anything needing escalation before the reply is sent.
Output format Sections: Appeal summary; Claim-by-claim assessment (table); Recommendation; Draft reply (under 200 words, plain language, no sarcasm, no internal jargon); Escalation note. Leave out speculation about the user's motives.
Guardrails
- Use only the rules, evidence and history supplied; never invent rule numbers, quotes, or prior incidents.
- Do not promise timelines, refunds, or outcomes not stated in the policy.
- Flag for escalation any threat of harm, legal claim, or policy gap, and say when a manager or trust and safety team must review before sending.
Example Platform: gaming forum; ban reason: harassment in chat; appeal: user says it was banter and a first offence; history: one prior warning.